ClawHub
Back to skill

Security audit

Auto Conda Env

Security checks across malware telemetry and agentic risk

Overview

The skill appears to perform disclosed local setup and dependency installation, with no evidence of hidden exfiltration, persistence, or destructive behavior.

Before installing, review the dependency commands and run them only in an isolated environment you control. Treat project dependency files as code, especially when using pip or conda, and avoid running installs against untrusted repositories.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill directs the agent to run state-changing package-management commands such as `conda create`, `conda env update`, and `pip install` based on project files, without an explicit user-consent checkpoint or a clear upfront warning that local system and environment state will be modified. In context, dependency files and package installation can execute untrusted build/install logic (`pip install .`, `setup.py`, package post-install/build steps), so automatic execution increases the risk of unintended code execution and environment tampering.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.